david baldus and the legacy of mccleskey v. kemp

University of Michigan Law SchoolUniversity of Michigan Law School Scholarship Repository

Articles Faculty Scholarship

2012

David Baldus and the Legacy of McCleskey v.KempSamuel R. GrossUniversity of Michigan Law School, [email protected]

Available at: https://repository.law.umich.edu/articles/182

Follow this and additional works at: https://repository.law.umich.edu/articles

Part of the Civil Rights and Discrimination Commons, Criminal Law Commons, and theSupreme Court of the United States Commons

This Article is brought to you for free and open access by the Faculty Scholarship at University of Michigan Law School Scholarship Repository. It hasbeen accepted for inclusion in Articles by an authorized administrator of University of Michigan Law School Scholarship Repository. For moreinformation, please contact [email protected].

Recommended CitationGross, Samuel R. "David Baldus and the Legacy of McCleskey v. Kemp." Iowa L. Rev. 97, no. 6 (2012): 1906-24.

https://repository.law.umich.edu?utm_source=repository.law.umich.edu%2Farticles%2F182&utm_medium=PDF&utm_campaign=PDFCoverPages

https://repository.law.umich.edu/articles?utm_source=repository.law.umich.edu%2Farticles%2F182&utm_medium=PDF&utm_campaign=PDFCoverPages

https://repository.law.umich.edu/faculty_scholarship?utm_source=repository.law.umich.edu%2Farticles%2F182&utm_medium=PDF&utm_campaign=PDFCoverPages

https://repository.law.umich.edu/articles/182

https://repository.law.umich.edu/articles?utm_source=repository.law.umich.edu%2Farticles%2F182&utm_medium=PDF&utm_campaign=PDFCoverPages

http://network.bepress.com/hgg/discipline/585?utm_source=repository.law.umich.edu%2Farticles%2F182&utm_medium=PDF&utm_campaign=PDFCoverPages



mailto:[email protected]

A4_GROSS.DOCX (DO NOT DELETE) 10/16/2012 11:12 AM

1905

David Baldus and the Legacy of McCleskey v. Kemp

Samuel R. Gross

I. INTRODUCTION .................................................................................... 1906

II. THE CONTEXT: RACIAL DISCRIMINATION AND THE

CONSTITUTIONAL REGULATION OF THE DEATH PENALTY IN THE

UNITED STATES .................................................................................... 1907

III. THE MCCLESKEY LITIGATION ............................................................... 1910 A. PRE-MCCLESKEY CASES ................................................................. 1910 B. MCCLESKEY IN THE LOWER COURTS .............................................. 1911 C. THE SUPREME COURT .................................................................... 1915

IV. THE AFTERMATH .................................................................................. 1917

V. CONCLUSION ....................................................................................... 1921

Thomas and Mabel Long Professor of Law, University of Michigan Law School.


1906 IOWA LAW REVIEW [Vol. 97:1905

I. INTRODUCTION

The first major empirical challenge to racial discrimination in the use of the death penalty in the United States was presented in federal court in the case of William L. Maxwell, who was sentenced to death in Arkansas in 1962 for the crime of rape.1 It was based on a landmark study by Marvin Wolfgang, a distinguished criminologist who had collected data on some 3000 rape convictions from 1945 through 1965 in selected counties across eleven southern states.2 He found that black men who were convicted of rape were seven times more likely to be sentenced to death than white men, and that black men who were convicted of raping white women were eighteen times more likely to be sentenced to death than men convicted of rape in any other racial combination.3 Wolfgang also examined other variables and found that the only one that was strongly related to death sentencing—the commission of a contemporaneous felony—did not explain these racial patterns.4

In 1968, the Eighth Circuit, in an opinion by Judge (later Justice) Harry Blackmun, rejected the Wolfgang study on three grounds.5 First, the court held that the data were not specific enough: too few cases came from the county in which Maxwell was prosecuted or even from Arkansas at all.6 Second, the data were not sufficiently detailed: “They admittedly do not take every variable into account.”7 Third, the study does not show intentional discrimination in Maxwell’s case: “They do not show that the petit jury which tried and convicted Maxwell acted in his case with racial discrimination.”8 Blackmun added:

We can understand and appreciate the disappointment and seeming frustration which Maxwell’s counsel must feel in again failing to prevail on a still more sophisticated statistical approach. They will ask themselves just how far they are required to go in order to prevail.

We are not certain that, for Maxwell, statistics will ever be his redemption. The facts as to rape charges in Garland County are known and have been recited. Standing by themselves, they

1. Maxwell v. Bishop, 257 F. Supp. 710 (E.D. Ark. 1966), aff’d, 398 F.2d 138 (8th Cir. 1968), vacated, 398 U.S. 262 (1970). 2. Marvin E. Wolfgang & Marc Riedel, Race, Judicial Discretion, and the Death Penalty, 407 AM. ACAD. POL. & SOC. SCI. 119, 127 (1973). 3. Id. at 129–30. 4. Id. at 132. 5. Maxwell v. Bishop, 398 F.2d 138 (8th Cir. 1968), vacated, 398 U.S. 262 (1970). 6. Id. at 146 7. Id. at 147 8. Id.


2012] BALDUS AND THE LEGACY OF MCCLESKEY V. KEMP 1907

disclose nothing from which conclusions of unconstitutionality in application may appropriately be drawn.9

The Supreme Court granted a writ of certiorari to review the Eighth Circuit’s decision in Maxwell and reversed on an unrelated issue without mentioning race.10

Nineteen years later, in McCleskey v. Kemp, the Supreme Court rejected another challenge to racial discrimination in the use of the death penalty.11 This time the state was Georgia and the challenge was based on a study by David Baldus. Justice Powell, writing for the Court, echoed some of Blackmun’s sentiments in Maxwell:

[McCleskey] offers no evidence specific to his own case that would support an inference that racial considerations played a part in his sentence. Instead, he relies solely on the Baldus study. McCleskey argues that the Baldus study compels an inference that his sentence rests on purposeful discrimination.

. . . .

[W]e hold that the Baldus study is clearly insufficient to support an inference that any of the decisionmakers in McCleskey’s case acted with discriminatory purpose.12

At first glance, it looks like nothing changed from 1968 to 1987. Indeed, since the Court has not revisited the issue since McCleskey, one could conclude that nothing changed from 1968 through the present. That would be a mistake. McCleskey was a turning point in the constitutional regulation of the death penalty in the United States, and it has influenced our collective view of race in the criminal-justice system generally. Its full impact is not yet known. The person most responsible for the decision in McCleskey was David Baldus.

In this Essay, I will briefly review the history and the enduring importance of the McCleskey case, and the work by David Baldus and his colleagues on which it was based.

II. THE CONTEXT: RACIAL DISCRIMINATION AND THE CONSTITUTIONAL

REGULATION OF THE DEATH PENALTY IN THE UNITED STATES

Racial discrimination has been the single most troubling issue for the death penalty in the United States in the past fifty years. It never goes away.

In 1965, the NAACP Legal Defense and Educational Fund (“Legal Defense Fund”), the law office that litigated Brown v. Board of Education13 and

9. Id. at 148. 10. Maxwell v. Bishop, 398 U.S. 262 (1970). 11. McCleskey v. Kemp, 481 U.S. 279 (1987). 12. Id. at 292–93, 297 (footnote omitted). 13. Brown v. Bd. of Educ., 347 U.S. 483 (1954).



many other civil rights cases, embarked on a systematic program of litigation in opposition to the death penalty. That decision led to the modern era of constitutional regulation of capital punishment. The Legal Defense Fund’s goal was, and remains, abolition of capital punishment for all defendants and for all crimes, but the reason that the civil rights organization took up the issue was their deep experience with racial discrimination in the use of the death penalty in the South. The Legal Defense Fund campaign against the death penalty produced Marvin Wolfgang’s landmark study and the unsuccessful litigation in Maxwell.14

In 1972, the Legal Defense Fund campaign against capital punishment culminated in Furman v. Georgia, in which a hopelessly fractured Supreme Court held that all then-existing death penalty statutes in the United States violated the Cruel and Unusual Punishment Clause of the Eighth Amendment.15 Furman is something of a Rorschach test. The five Justices in the majority wrote five separate one-Justice opinions. Furman is understood to prohibit the “arbitrary” imposition of the death penalty. That is a rough description of the positions of Justices Stewart and White, the two Justices who joined the majority on the narrowest grounds.16 More important, that is how Furman is described in later opinions of the Court.17

In an alternate universe, Furman might have marked the end of capital punishment in the United States. But the Court did not say that death is an inherently cruel and unusual punishment, and dozens of states responded to Furman by enacting new death-sentencing laws that attempted to remedy the problem of “arbitrariness.” In 1976, in Gregg v. Georgia, the Court held that several of those new statutes were at least potentially constitutional—those that, like Georgia’s new death-penalty law, provided for “guided discretion” to juries and judges in imposing death sentences and therefore, presumably, reduced or eliminated the “arbitrariness” that was condemned in Furman.18

Since Gregg, racial discrimination in the use of the death penalty has been understood to be unconstitutional under the Cruel and Unusual Punishments Clause of the Eighth Amendment as well as the Equal Protection Clause of the Fourteenth Amendment. Three of the Justices in the Furman majority discussed discrimination in capital sentencing,19 and at

14. See generally MICHAEL MELTSNER, CRUEL AND UNUSUAL: THE SUPREME COURT AND

CAPITAL PUNISHMENT (1973). 15. Furman v. Georgia, 408 U.S. 238 (1972). 16. Id. at 306–10 (Stewart, J., concurring); id. at 310–14 (White, J., concurring). 17. Gregg v. Georgia, 428 U.S. 153, 188 (1976). 18. See id. at 188–95; see also Roberts v. Louisiana, 431 U.S. 633 (1977); Proffitt v. Florida, 428 U.S. 242 (1976); Jurek v. Texas, 428 U.S. 262 (1976); Woodson v. North Carolina, 428 U.S. 280 (1976). 19. Furman, 408 U.S. at 255–57 (Douglas, J., concurring); id. at 310 (Stewart, J., concurring); id. at 364–65 (Marshall, J., concurring).



least two of them seemed to rely on it in reaching their separate judgments.20 In any event, the received wisdom after 1976 was that racial discrimination is an element of the “arbitrariness” that “was condemned in Furman” and one of the evils that the post-Furman capital-sentencing reforms were designed to cure.21

In the wake of Gregg, the Supreme Court has been mired in an endless, contentious, and sometimes bizarre program of constitutional regulation of the death penalty.22 I will not begin to discuss that huge and confusing topic, except to note that in general it addresses procedure: What rules are permitted or required for a constitutional death penalty? That gives the Court a great deal of freedom. When the question is, for example, the constitutionality of executing defendants for crimes committed before they reached eighteen years of age,23 the Court can both define the issues and set the rules.

Race discrimination is different. No one doubts that racial discrimination by the state is unconstitutional, in this and in almost every other context. The question is actual practice: Did the state in fact discriminate by race in imposing and executing death sentences? It is concrete and potentially incendiary. Courts can define and redefine the term “discrimination” to produce the outcome they want, but they have to deal with facts on the ground, which can be ugly.

The Supreme Court has almost never found systemic racial discrimination in the administration of criminal justice by the states. In 1996, in United States v. Armstrong, the Court claimed that its requirements for proof of racial discrimination under the Equal Protection Clause “[do] not make a selective-prosecution claim impossible to prove”24—but the only successful claim it could cite was in Yick Wo v. Hopkins,25 which had been decided in 1886, 110 years earlier. In most cases, the Court simply ducks the issue. That’s what the Court did in Maxwell in 1968. In 1977, one year after Gregg, it did it again. In Coker v. Georgia, the Supreme Court held that the death penalty is unconstitutional for the crime of rape.26 The Court, of course, was well aware of the notorious racist history of capital punishment for rape. The Wolfgang study was presented to it in Maxwell v. Bishop27 and

20. See id. at 255–57 (Douglas, J., concurring); id. at 364–65 (Marshall, J., concurring). 21. See, e.g., McCleskey v. Kemp, 753 F.2d 877, 890–91 (11th Cir. 1985), aff’d, 481 U.S. 279 (1987); Spinkellink v. Wainwright, 578 F.2d 582, 613–14 & n.38 (5th Cir. 1978). 22. See, e.g., Carol S. Steiker & Jordan M. Steiker, Sober Second Thoughts: Reflections on Two Decades of Constitutional Regulation of Capital Punishment, 109 HARV. L. REV. 355 (1995); Robert Weisberg, Deregulating Death, 1983 SUP. CT. REV. 305 (1983). 23. Roper v. Simmons, 543 U.S. 551 (2005). 24. United States v. Armstrong, 517 U.S. 456, 466 (1996). 25. Yick Wo v. Hopkins, 118 U.S. 356, 373 (1886). 26. Coker v. Georgia, 433 U.S. 584, 592 (1977). 27. Maxwell v. Bishop, 398 U.S. 262 (1970).



was discussed in briefs in later cases, including Furman and Gregg. Nonetheless, the Court’s opinion in Coker makes no mention of race. If you didn’t know better, you could read it and never realize that there is more than one race in the United States.

III. THE MCCLESKEY LITIGATION

A. PRE-MCCLESKEY CASES

After Gregg, it was only a matter of time before the question of racial discrimination in the use of the death penalty would surface again. The earliest challenges were based on research that could be done relatively quickly and inexpensively. In particular, a 1980 study by William Bowers and Glenn Pierce compared reported homicides and death sentences in Florida, Georgia, Texas, and Ohio.28 In each state they found that courts were more likely to impose death sentences for homicides with white victims than for those with black victims and that black defendants charged with killing white victims were more likely to be sentenced to death than white defendants charged with killing white victims.29

The lower federal courts rejected these challenges by the simplest means available. In 1981, for example, in Smith v. Balkcom, the Fifth Circuit held that because the Bowers and Pierce study left “untouched countless racially neutral variables”30—variables that describe the charging of reported homicides, the disposition of those charges at trial, and the presence of aggravating and mitigating factors that bear on the choice of punishment—it could be ignored, without a hearing to explore its validity and significance.31

By 1982, David Baldus and his colleagues had completed the first of their two major studies of death sentencing in Georgia, the Procedural Reform Study,32 and three prisoners on Georgia’s death row offered it in a federal habeas corpus proceeding in support of their joint claim of discrimination in capital sentencing. In a supplemental opinion issued in light of Baldus’s study, the district court, tracking the circuit court’s language in Smith but

28. William J. Bowers & Glenn L. Pierce, Arbitrariness and Discrimination Under Post-Furman Capital Statutes, 26 CRIME & DELINQ. 563 (1980). 29. Id. at 594 tbl.2. 30. Smith v. Balkcom, 671 F.2d 858, 859 (5th Cir. 1982). 31. Id. at 860; see also Spinkellink v. Wainwright, 578 F.2d 582, 616 (5th Cir. 1978) (rejecting an equal protection challenge because evidence put forth “could not prove discriminatory intent or purpose”). 32. For a discussion of this study, see McCleskey v. Zant, 580 F. Supp. 338, 353–55 (N.D. Ga. 1984), rev’d in part, aff’d in part sub nom. McCleskey v. Kemp, 753 F.2d 877 (11th Cir. 1985), aff’d, 481 U.S. 279 (1987).



ignoring the record before it, dismissed the claim because Baldus’s study “leaves untouched countless racially neutral variables.”33

That was a mistake. The hallmark of David Baldus’s work was his commitment to leave no case and no variable untouched. When Spencer v. Zant, the first of the three cases reached the Eleventh Circuit on appeal, it was reversed and remanded for an evidentiary hearing: “Dr. Baldus’s study [may have] addressed the very defects identified in the evidence in . . . Smith . . . . The merits of this allegation cannot be assessed without a more detailed consideration of the evidence proffered by the petitioners below.”34

B. MCCLESKEY IN THE LOWER COURTS

By the time Spencer was decided on appeal in September 1983, both Baldus studies—the more comprehensive Charging and Sentencing Study as well as the Procedural Reform Study—had been completed and presented in another case. Warren McCleskey, a black man, was sentenced to death in Fulton County, Georgia, in 1978 for killing a white police officer. In late 1981, Legal Defense Fund lawyers representing McCleskey filed a petition for a writ of habeas corpus in the United States District Court for the Northern District of Georgia, claiming, among other issues, that his death sentence was the product of racial discrimination, as shown by the Baldus research. In 1984, the Eleventh Circuit stayed rehearing proceedings in Spencer pending the outcome of the same claim in McCleskey’s case,35 which had become the designated vehicle for consideration of the Baldus studies in the federal courts.

Before David Baldus, studies of racial discrimination in the use of the death penalty were difficult but manageable undertakings. Some studies used data that had been compiled by government agencies, especially the FBI’s Uniform Crime Reports.36 Others—even Marvin Wolfgang’s impressive path-breaking work—sampled comparatively small numbers of cases and considered limited numbers of non-racial variables that might explain any racial disparities. The research conducted by Baldus and his colleagues was different in kind. The basic posture—which Dave Baldus personified—was simple: “Why not find out everything about every case?”

For most of us, that is not a rhetorical question. The unfortunate answer is that we have limited time, money, and energy. Baldus was subject to two of those limits—time and money—but did more with what he had than seems humanly possible. There is no empirical evidence that he ever

33. Ross v. Hopper, 538 F. Supp. 105, 107 (S.D. Ga. 1982) (quoting Balkcom, 671 F.2d at 859), modifying Mitchell v. Hopper, 538 F. Supp. 77 (S.D. Ga. 1982), aff’d in part, remanded in part en banc sub nom. Ross v. Kemp, 756 F.2d 1483 (11th Cir. 1985). 34. Spencer v. Zant, 715 F.2d 1562, 1582 (11th Cir. 1983), reh’g granted en banc sub nom. Spencer v. Kemp, 781 F.2d 1458 (11th Cir. 1986). 35. Spencer v. Zant, 729 F.2d 1293, 1294 (11th Cir. 1984). 36. E.g., Bowers & Pierce, supra note 28, at 591.



lacked energy. The Georgia studies that were the subject of the McCleskey litigation are prime examples of his style of work.

For the Procedural Reform Study, David Baldus and his colleagues, Charles Pulaski and George Woodworth, compiled data on over 200 variables for 594 defendants who were tried and sentenced for murder in Georgia from March 1973 through July 1978.37 This was an uncommonly comprehensive sentencing study, but it had limitations. It lacked data on the strength of the evidence of the defendant’s guilt, and, since it was restricted to murder convictions, it did not examine the possibility of pretrial discrimination in charging and plea bargaining. These gaps were filled by the Charging and Sentencing Study, which covered 1066 Georgia homicide prosecutions from 1973 through 1980, manslaughter convictions and guilty pleas as well as murder convictions, and included detailed data on an expanded list of over 400 nonracial variables.38 The court opinions in McCleskey focus on the more comprehensive Charging and Sentencing Study. (In fact, they uniformly refer to the Baldus “study” despite the fact that both Baldus studies are in the record. I will follow suit and refer to the Baldus research in the McCleskey record as a single study.)

I will not try to summarize the findings of the Baldus study. A small library has been published on the topic, including a book by Baldus himself with George Woodworth and Charles Pulaski.39 Suffice it to say that Baldus and his colleagues found a strong and consistent pattern of discrimination in the use of the death penalty against defendants who were charged with killing white victims compared to those who were charged with killing black victims. They also found a weaker pattern of discrimination against black defendants (for example, Warren McCleskey himself) in homicide cases in which the victim was white. Finally, the study had a sufficient number of cases and sufficiently detailed data to show that these patterns applied to homicides in Fulton County, where McCleskey was convicted and sentenced to death, as well as in Georgia as a whole.

Like other courts that faced this issue, the district court rejected McCleskey’s claims of racial discrimination in capital sentencing on empirical grounds.40 Given the Eleventh Circuit’s opinion in Spencer, however, the judge was unable to do so on the preferred basis: failure to examine a sufficient number of nonracial variables. Instead, he attacked the Baldus study on several other fronts: (1) the database was too inaccurate to

37. See McCleskey, 580 F. Supp. at 353–55. 38. See id. 39. DAVID C. BALDUS, GEORGE WOODWORTH & CHARLES A. PULASKI, JR., EQUAL JUSTICE

AND THE DEATH PENALTY: A LEGAL AND EMPIRICAL ANALYSIS (1990). 40. McCleskey, 580 F. Supp. at 346–80. The district court did grant an unrelated claim by McCleskey and ordered a new trial. Id. at 384. That order was reversed by the Eleventh Circuit on appeal. McCleskey v. Kemp, 753 F.2d 877 (11th Cir. 1985), aff’d, 481 U.S. 279 (1987).



form a basis for useful conclusions;41 (2) the statistical models used by Baldus and his colleagues were flawed;42 (3) the data did not demonstrate that the capital-sentencing system in Georgia was discriminatory;43 and (4) the statistical methodology used had no value in this context.44 In short, the judge concluded that the Baldus study “fail[ed] to contribute anything of value” to McCleskey’s claim.45

The district court’s opinion in McCleskey includes thirty-five pages on the claim of racial discrimination.46 Its discussion of the Baldus study and of statistics is detailed and self-assured—and probably pretty convincing, unless you happen to know something about the record in the case or about statistics. In fact, most of the criticisms of Professor Baldus’s research are unfair and inaccurate, and many of the statements about statistics are simply false, as I have discussed at length elsewhere.47 But there is little reason to pay attention to the district court opinion. Its rationale and conclusions were all but ignored by the Eleventh Circuit on appeal and by the Supreme Court in its review of the Eleventh Circuit.

The Eleventh Circuit opinion in McCleskey notes that “[t]he district court held the [Baldus] study to be invalid,”48 but the circuit court itself takes a different tack.

We assume without deciding that the Baldus study is sufficient to show . . . . that systematic and substantial disparities existed in the penalties imposed upon homicide defendants in Georgia based on race of the homicide victim, that the disparities existed at a less substantial rate in death sentencing based on race of defendants, and that the factors of race of the victim and defendant were at work in Fulton County.49

The court “pretermit[s] a review of . . . the validity of the study itself” because “even if the statistical results are accepted as valid, the evidence fails to challenge successfully the constitutionality of the Georgia system.”50

The problem for the circuit court is the magnitude of the racial discrimination found by the study. The court holds that no hearing or factual assessment is required for a statistical study of discrimination in capital sentencing, regardless of its quality, unless it “reflect[s] a [racial]

41. McCleskey, 580 F. Supp. at 354–60. 42. Id. at 360–64. 43. Id. at 364–69, 372–77. 44. Id. at 369–72. 45. Id. at 372. 46. Id. at 346–80. 47. See, e.g., SAMUEL R. GROSS & ROBERT MAURO, DEATH & DISCRIMINATION: RACIAL

DISPARITIES IN CAPITAL SENTENCING 153 nn.20 & 21 (1989). 48. McCleskey v. Kemp, 753 F.2d 877, 894 (11th Cir. 1985), aff’d, 481 U.S. 279 (1987). 49. Id. at 895. 50. Id. at 894–95.



disparity so great as to inevitably lead to a conclusion that the disparity results from intent or motivation.”51 The Baldus study fails that test. “The result of Baldus’ [sic] most conclusive model, on which McCleskey primarily relies, showed an effect of 0.06, signifying that on average a white victim crime is 6% more likely to result in the [death] sentence than a comparable black victim crime.”52 This “6% bottom line” is “not sufficient to overcome the presumption that the statute is operating in a constitutional manner.”53

The circuit court holding in McCleskey is hard to fathom. “Six percent” may not seem like a lot in some contexts—say the difference between male and female employment at a plant with 53% men and 47% women. On the other hand, as I write in 2012, 6% interest on a savings account would be astronomically high, and a 6% annual growth rate in the gross domestic product is a Utopian dream. In the context of capital sentencing in Georgia in the 1970s, a good description of the “6%” racial disparity found by Baldus (after controlling for many other variables) is that it corresponds to an increase in the probability of a death sentence from 3% to 9%.54 Did the Eleventh Circuit really mean to say that an unexplained racial disparity that increases the risk of execution by a factor of three is just too small to require consideration?

But why did the Eleventh Circuit choose this peculiar justification? Why not follow the district court and reject the study on methodological grounds? The common method for an appellate court to affirm a trial court decision that is based on an elaborate record is to endorse the factual findings of the trial judge. After all, trial courts are supposed to do the heavy lifting in evaluating facts, and their decisions are entitled to a great deal of deference.55 Why didn’t the Eleventh Circuit take that simple, easy route? The answer, I believe, is that the circuit court judges realized that the Baldus study could not be dismissed so easily. The McCleskey case was heading to the Supreme Court, and the Baldus study had already attracted a great deal of favorable attention. As a dissenting judge pointed out, it was already described in the record by a distinguished statistician and criminologist as “far and away the most complete and thorough analysis of sentencing” ever conducted.56 The circuit court may have been unwilling to rest its judgment on the untenable claim that the most thorough study of sentencing patterns

51. Id. at 894 (emphasis added). 52. Id. at 896. 53. Id. at 897. 54. GROSS & MAURO, supra note 47, at 147. 55. See, e.g., FED. R. CIV. P. 52(a)(6) (“Findings of fact, whether based on oral or other evidence, must not be set aside unless clearly erroneous, and the reviewing court must give due regard to the trial court’s opportunity to judge the witnesses’ credibility.”). 56. See McCleskey, 753 F.2d at 907 (Johnson, J., concurring in part and dissenting in part) (quoting the description of Dr. Robert Berk at trial).



ever conducted in this country was inadequate to satisfy its methodological demands.

C. THE SUPREME COURT

The Supreme Court affirmed the Eleventh Circuit’s decision in McCleskey by a five-to-four vote, with Justice Powell writing for the majority.57 After reviewing the history of the case, Powell quickly makes clear that he will not address any questions of fact: “As did the Court of Appeals, we assume the [Baldus] study is valid statistically without reviewing the factual findings of the District Court.”58 Instead, he rejects McCleskey’s claims because “[h]e offers no evidence specific to his own case that would support an inference that racial considerations played a part in his sentence.”59

As I mentioned in the Introduction, this holding echoes the 1968 decision on William Maxwell’s claim of racial discrimination in capital sentencing for rape—but only in part. The Eighth Circuit rejected Maxwell’s claim on two additional grounds, both empirical: Marvin Wolfgang’s study was also held to be inadequate because it included too few cases from the jurisdiction in question and too few nonracial control variables.60 In McCleskey, the Court backs away from any empirical criticism of the Baldus study. Instead it relies exclusively on the third basis for the Maxwell decision, the legal requirement that to prove discrimination in capital sentencing a defendant must produce specific evidence that the decision makers in his own individual case acted with a racially discriminatory purpose.

One of the striking aspects of the succession of opinions in McCleskey is the progressive evaporation of the factual question with which the case began: Does the Baldus study prove race discrimination in capital sentencing in Georgia? The district court took on the study directly and held that it is so flawed that it proves nothing. The court of appeals retreated, but only halfway: it assumed that the study was valid but rejected it on the inexplicable empirical ground that the magnitude of discrimination shown was constitutionally insufficient. The Supreme Court eliminated all empirical issues entirely by deciding that this type of evidence cannot in principle establish a violation of the Constitution.

Why this increasing empirical modesty as the case moved up the judicial ladder? Was it because at each higher step, as the record received more attention, it became increasingly clear that the Baldus study could not be

57. McCleskey v. Kemp, 481 U.S. 279 (1987). 58. Id. at 291 n.7. 59. Id. at 292–93. 60. Maxwell v. Bishop, 398 F.2d 138, 146–47 (8th Cir. 1968), vacated, 398 U.S. 262 (1970).



rejected on its own terms,61 and increasingly attractive to retreat to the safe turf of categorical legal rules?

The meaning of the majority’s factual concession in McCleskey is driven home by one of the three dissenting opinions. Justice Brennan, joined by Justices Marshall, Blackmun, and Stevens, describes a concrete consequence of the system the majority upholds, as it is described in the Baldus study:

At some point in this case, Warren McCleskey doubtless asked his lawyer whether a jury was likely to sentence him to die. A candid reply to this question would have been disturbing. First, counsel would have to tell McCleskey that few of the details of the crime or of McCleskey’s past criminal conduct were more important than the fact that his victim was white. . . . In addition, frankness would compel the disclosure that it was more likely than not that the race of McCleskey’s victim would determine whether he received a death sentence . . . . Finally, the assessment would not be complete without the information that cases involving black defendants and white victims are more likely to result in a death sentence than cases featuring any other racial combination of defendant and victim. The story could be told in a variety of ways, but McCleskey could not fail to grasp its essential narrative line: there was a significant chance that race would play a prominent role in determining if he lived or died.62

Justice Blackmun’s dissent, for the four justices who also signed Brennan’s opinion, describes the distance travelled from Maxwell to McCleskey in detailed and personal terms:

As a member of the United States Court of Appeals, I was confronted in 1968 with a challenge to the constitutionality of a State’s capital sentencing system based on allegations of racial discrimination supported by statistical evidence. Writing for a panel of the court [in Maxwell v. Bishop], I rejected that challenge for reasons similar to those espoused by the Court today.

. . . .

The Court of Appeals found the evidence presented by Maxwell incomplete, not directly relevant to his individual claim, and statistically insufficient. McCleskey’s evidence, however, is of such a different level of sophistication and detail that it simply cannot be

61. A brief filed in the Supreme Court by several of the country’s preeminent criminologists described the Baldus study as “among the best empirical studies on criminal sentencing ever conducted.” Brief for Dr. Franklin M. Fisher, Dr. Richard O. Lempert, Dr. Peter W. Sperlich, Dr. Marvin E. Wolfgang, Professor Hans Zeisel and Professor Franklin E. Zimring as Amici Curiae Supporting Petitioner at 3, McCleskey, 481 U.S. 279 (No. 84-6811). 62. McCleskey, 481 U.S. at 322–23 (Brennan, J., dissenting) (citations omitted).



rejected on those grounds. Unlike the evidence presented by Maxwell, which did not contain data from the jurisdiction in which he was tried and sentenced, McCleskey’s evidence includes data from the relevant jurisdiction. Whereas the analyses presented by Maxwell did not take into account a significant number of variables and were based on a universe of 55 cases, the analyses presented by McCleskey’s evidence take into account more than 400 variables and are based on data concerning all offenders arrested for homicide in Georgia from 1973 through 1978, a total of 2,484 cases. Moreover, the sophistication of McCleskey’s evidence permits consideration of the existence of racial discrimination at various decision points in the process, not merely at the jury decision. It is this experience, in part, that convinces me of the significance of the Baldus study.63

But the true significance of the McCleskey decision as a factual judgment is best conveyed by Justice Powell’s majority opinion, which speaks for the Court. A couple of incidental points at the end of the Court’s opinion are telling. They look like make-weight arguments, but they offer a window into Justice Powell’s thinking. McCleskey, he says, makes “wide-ranging arguments that basically challenge the validity of capital punishment in our multiracial society,” but the Court can only decide “whether in his case . . . the law of Georgia was properly applied.”64 Courts don’t deal with systemic claims like these for two reasons. First “there is no limiting principle to the type of challenge brought by McCleskey”—they could apply to sentences other than death and to discrimination by ethnicity or gender as well as by race.65 Second, “[l]egislatures . . . are better qualified to weigh and ‘evaluate the results of statistical studies in terms of their own local conditions and with a flexibility of approach that is not available to the courts.’”66

In other words, Justice Powell seems to say, “It does look like there’s a real problem here. We don’t deny it. But we’re not equipped to help you. Ask elsewhere.”

IV. THE AFTERMATH

McCleskey remains one of the most controversial decisions in the history of the Supreme Court. It is often compared to other notorious cases in which the Court endorsed discrimination by race67: Dred Scott v. Sandford,68

63. Id. at 354 n.7 (Blackmun, J., dissenting) (citation omitted). 64. Id. at 319 (majority opinion) (citation omitted). 65. Id. at 318. 66. Id. at 319 (quoting Gregg v. Georgia, 428 U.S. 153, 186 (1976)). 67. Randall L. Kennedy, McCleskey v. Kemp: Race, Capital Punishment, and the Supreme Court, 101 HARV. L. REV. 1388, 1388–89 (1988). 68. Dred Scott v. Sandford, 60 U.S. (19 How.) 393 (1857).



Plessy v. Ferguson,69 and Korematsu v. United States.70 Six days after it was decided, Anthony Lewis wrote in the New York Times that the Court had “effectively condoned the expression of racism in a profound aspect of our law.”71 Decades later, the criticism continues.72

In 1990, the Congressional Black Caucus responded to McCleskey by introducing the Racial Justice Act, which provided that “no person shall be put to death under color of State or Federal law in the execution of a sentence that was imposed based on race,” and permitted courts to infer racial discrimination from statistical evidence.73 The Act was passed by the United States House of Representatives as part of crime legislation packages in 1990 and again in 1994, but it was deleted from the legislation each time in conference with the United States Senate.74 In 1998, a weak Racial Justice Act was signed into law in Kentucky.75 In 2009 North Carolina passed a much stronger Racial Justice Act,76 which is the basis for ongoing litigation on racial discrimination in capital sentencing in that state.77

The most telling responses to McCleskey, however, have come from Justices of the Supreme Court.

Justice Powell retired in June 1987, two months after he wrote the opinion of the Court in McCleskey. Three years later, his biographer asked him whether he would change his vote in any case. He replied:

“Yes, McCleskey v. Kemp.” “Do you mean you would now accept the argument from statistics?” “No, I would vote the other way in any capital case.” “In any capital case?” “Yes.” “Even in Furman v. Georgia?”

69. Plessy v. Ferguson, 163 U.S. 537 (1896). 70. Korematsu v. United States, 323 U.S. 214 (1944). 71. Anthony Lewis, Bowing to Racism, N.Y. TIMES, Apr. 28, 1987, at A31. 72. See, e.g., Paul Butler, By Any Means Necessary: Using Violence and Subversion To Change Unjust Law, 50 UCLA L. REV. 721, 730–33 (2003). 73. Don Edwards & John Conyers, Jr., The Racial Justice Act—A Simple Matter of Justice, 20 U. DAYTON L. REV. 699, 700–01, 704 (1995). 74. See id. at 700–01. The main argument against the Racial Justice Act was that it would amount to an abolition of the death penalty. See Daniel E. Lungren & Mark L. Krotoski, The Racial Justice Act of 1994—Undermining Enforcement of the Death Penalty Without Promoting Racial Justice, 20 U. DAYTON L. REV. 655, 655 (1995). 75. 1998 Ky. Acts 941 (codified at Kentucky Racial Justice Act, KY. REV. STAT. ANN. §§ 532.300–09 (West 2006)). The Kentucky Racial Justice Act is prospective only—it applies only to defendants sentenced to death after July 15, 1998—and is limited to pre-trial claims of racial discrimination in capital charging. See Justin R. Arnold, Note, Race and the Death Penalty After McCleskey: A Case Study of Kentucky’s Racial Justice Act, 12 WASH. & LEE J. CIVIL RTS. & SOC. JUST. 93, 102–03 (2005). 76. 2009 N.C. Sess. Laws 1213 (codified at North Carolina Racial Justice Act, N.C. GEN. STAT. § 15A-2010 (2011)). 77. See, e.g., Editorial, Race and Death Penalty Juries, N.Y. TIMES, Feb. 6, 2012, at A22.



“Yes. I have come to think that capital punishment should be abolished.”78

Three years after that, in February 1994—six months before he too retired—Justice Blackmun issued an opinion, dissenting from the denial of certiorari in the death penalty case of Callins v. Collins,79 in which he announced that he had concluded that the death penalty is unconstitutional:

From this day forward, I no longer shall tinker with the machinery of death. . . . Rather than continue to coddle the Court’s delusion that the desired level of fairness has been achieved and the need for regulation eviscerated, I feel morally and intellectually obligated simply to concede that the death penalty experiment has failed.80

One of the main reasons that Blackmun cites for this change of heart is the Baldus study:

A renowned example of racism infecting a capital-sentencing scheme is documented in McCleskey v. Kemp . . . . Warren McCleskey, an African-American, argued that the Georgia capital sentencing scheme was administered in a racially discriminatory manner, in violation of the Eighth and Fourteenth Amendments. In support of his claim, he proffered a highly reliable statistical study (the Baldus study) which indicated that, “after taking into account some 230 nonracial factors that might legitimately influence a sentencer, the jury more likely than not would have spared McCleskey’s life had his victim been black[,]” . . . . [and] that blacks who kill whites are sentenced to death “at nearly 22 times the rate of blacks who kill blacks, and more than 7 times the rate of whites who kill blacks.”

. . . .

. . . [A]s far as I know, there has been no serious effort to impeach the Baldus study. Nor, for that matter, have proponents of capital punishment provided any reason to believe that the findings of that study are unique to Georgia.81

In April 2008, Justice Stevens announced in a concurring opinion in Baze v. Rees that he too had concluded that the death penalty is unconstitutional.82 One of the reasons Stevens cites for his change of heart is

78. JOHN C. JEFFERIES, JR., JUSTICE LEWIS F. POWELL, JR. 451 (1994). 79. Callins v. Collins, 510 U.S. 1141 (1994). 80. Id. at 1145 (Blackmun, J., dissenting). 81. Id. at 1153–54 (citations omitted). 82. Baze v. Rees, 553 U.S. 35, 86 (2008) (Stevens, J., concurring).



the “discriminatory application of the death penalty,” as shown by the record in McCleskey.83 Two years later, not long after his retirement in June 2010, Stevens told an interviewer that the only vote he regretted in his thirty-five years on the Court was his 1976 vote in Gregg v. Georgia.84 The votes of Justices Stevens, Powell, and Blackmun were essential to the seven-to-two decision in Gregg to uphold the constitutionality of the death penalty. By the time they were persuaded otherwise it was too late to change that outcome.

Other Justices have also cited the Baldus study, and the McCleskey case in general, to make the point that the administration of the death penalty is infected by racial discrimination.85 This is not surprising. Supreme Court Justices cite lots of sources. What’s more telling is that these claims are never disputed. Justice Scalia in particular is absent from this debate. On other issues, he frequently takes it upon himself to respond personally to other Justices who express concerns about capital punishment. He wrote separate concurring opinions in Callins v. Collins and in Baze v. Rees to rebut the opinions in which Justices Blackmun and Stevens, respectively, announced their conclusions that the death penalty is unconstitutional. But nowhere in those opinions, or elsewhere, does Justice Scalia say anything about racial discrimination.

Justice Scalia is certainly not averse to judicial conflict, in this or other contexts. In the case of Kansas v. Marsh, for example, Justice Souter, joined by Justices Stevens, Ginsburg, and Breyer, wrote a dissenting opinion in which five pages were devoted to their anxiety about the dangers of false convictions and the large number of recent exonerations of death row prisoners in the United States.86 Scalia responded with fourteen furious pages in which he belittles their concerns, ridicules the studies they discuss, and endorses the absurd claim that the American system of criminal adjudication is “99.973 percent” accurate.87

Why then this uncharacteristic shyness when the issue is racial discrimination?

After the death of Justice Thurgood Marshall in 1993, the Library of Congress made his papers public. They include a one paragraph memo by Justice Scalia to the other Justices, dated January 6, 1987, about the McCleskey case:

83. Id. at 85. 84. Nina Totenberg, Justice Stevens: An Open Mind on a Changed Court, NPR (Oct. 4, 2010), http://www.npr.org/templates/story/story.php?storyId=130198344. 85. E.g., Ring v. Arizona, 536 U.S. 584, 617 (2002) (Breyer, J., concurring). 86. Kansas v. Marsh, 548 U.S. 163, 207–11 (2006) (Souter, J., dissenting). 87. Id. at 182–99, 198 (Scalia, J., concurring) (quoting Joshua Marquis, Op-Ed., The Innocent and the Shammed, N.Y. TIMES, Jan. 26, 2006, at A23).



Re: No. 84-6811—McCleskey v. Kemp

MEMORANDUM TO THE CONFERENCE:

I plan to join Lewis’s [Justice Lewis Powell’s] opinion in this case, with two reservations. I disagree with the argument that the inferences that can be drawn from the Baldus study are weakened by the fact that each jury and each trial is unique, or by the large number of variables at issue. And I do not share the view, implicit in the opinion, that an effect of racial factors upon sentencing, if it could only be shown by sufficiently strong statistical evidence, would require reversal. Since it is my view that unconscious operation of irrational sympathies and antipathies, including racial, upon jury decisions and (hence) prosecutorial decisions is real, acknowledged in the decisions of this court, and ineradicable, I cannot honestly say that all I need is more proof. I expect to write separately to make these points, but not until I see the dissent.88

In other words, Scalia was persuaded in the McCleskey case itself that Baldus had proved racial discrimination in the administration of the death penalty in Georgia. Since then he has apparently followed the polite precept that if you don’t have something nice to say about the death penalty, don’t say anything.89

V. CONCLUSION

I don’t want to sound Pollyannaish. McCleskey remains the law, and it is a terrible decision. Race discrimination in the administration of the death penalty continues. But however bad, that’s not the whole picture.

In addition to reporting that he regretted his decision in McCleskey, Justice Powell told his biographer, “[m]y understanding of statistical analysis . . . ranges from limited to zero.”90 Powell was probably telling the truth about statistics. He may have had no clue what the Baldus study really meant on its own terms. But he had no difficulty understanding what was at stake in McCleskey. The question in McCleskey was indeed, as Powell phrased it years later, whether “capital punishment should be abolished.”91

In Furman, the Court nullified all then-existing death penalty statutes. It wiped the slate clean. In Gregg, the Court gave states a chance to try to

88. Erwin Chemerinsky, Eliminating Discrimination in Administering the Death Penalty: The Need for the Racial Justice Act, 35 SANTA CLARA L. REV. 519, 528 (1995). 89. See EDWARD P. LAZARUS, CLOSED CHAMBERS 211 (1998) (concluding based on various sources inside the Court that Justice Scalia was persuaded by the Baldus study but was “willing to tolerate that bias and even thought that the other Justices, in candor, should admit that they were too”) 90. JEFFERIES, supra note 78, at 439. 91. See id.; supra text accompanying note 78.



administer their new, post-Furman death penalties fairly. By 1987, Georgia’s statute had been in effect for fourteen years. What would the Court have done if it had concluded that in practice the new post-Furman Georgia statute was permeated by unconstitutional discrimination? It could hardly have said: “Well, that didn’t work, so let’s start over again, and try, again, to create a fair system. We’ll check in ten or fifteen years from now and see if this time you get it right.” As Powell and his colleagues apparently recognized, if McCleskey had prevailed, the Supreme Court would probably have had to abolish capital punishment, in chunks if not in one blow. In 1987, the homicide rate and public support for the death penalty were both very high, and had been for years.92 The Court was not about to do any such thing.

The present Supreme Court is not about to abolish capital punishment either, but the surrounding climate has changed. Public support for the death penalty has decreased sharply over the past fifteen years.93 Five states have abolished capital punishment since 200794 and others may soon follow. In 2011, the number of new sentences was about a quarter of what it was fifteen years earlier,95 and the number of executions was half that in 2000.96 The death penalty in the United States is on the decline.

Concern about racial discrimination is not the driving force behind this loss of enthusiasm for executions. The main reasons appear to be a rapid decline in the crime rate, especially the homicide rate, beginning in the early 1990s, and rising anxiety about the danger of executing innocent defendants.97 In this new environment, however, the issue of racism in the use of the death penalty has gained power; in 2009, for example, it produced the Racial Justice Act in North Carolina.98

The main reason that race is a powerful issue in debates about the death penalty is that everyone who cares knows that race plays a major role in determining who gets sentenced to death. And the single most important

92. See generally Samuel R. Gross & Phoebe C. Ellsworth, Hardening of the Attitudes: Americans’ Views on the Death Penalty, 50 J. SOC. ISSUES 19 (1994). 93. Death Penalty, GALLUP, http://www.gallup.com/poll/1606/Death-Penalty.aspx (last visited Feb. 13, 2012). 94. States with and Without the Death Penalty, DEATH PENALTY INFO. CTR., http://www. deathpenaltyinfo.org/states-and-without-death-penalty (last visited May 20, 2012) (noting that Connecticut, New York, New Jersey, New Mexico, and Illinois have abolished the death penalty since 2007). 95. DEATH PENALTY INFO. CTR., THE DEATH PENALTY IN 2011: YEAR END REPORT 3 (2011), available at http://www.deathpenaltyinfo.org/documents/2011__Year__End.pdf. 96. Id. at 1. 97. See generally Samuel R. Gross & Phoebe C. Ellsworth, Second Thoughts: Americans’ Views on the Death Penalty at the Turn of the Century, in BEYOND REPAIR? AMERICA'S DEATH PENALTY (Stephen P. Garvey ed., 2003). 98. 2009 N.C. Sess. Laws 1213 (codified at North Carolina Racial Justice Act, N.C. GEN. STAT. § 15A-2010 (2011)).



reason that “everyone knows” this is what happened in McCleskey. Even on the Supreme Court that sent Warren McCleskey to his death, even among the Justices who most strongly support the death penalty, nobody has tried to deny that racial “sympathies and antipathies” decide who lives and who dies. No Justice said otherwise in McCleskey and none have denied it since. That may be the enduring legacy of McCleskey.

But didn’t everybody know this all along? Weren’t the facts on the ground undeniable? I don’t think so. Facts are “undeniable” once we’re all convinced they’re true. If the Baldus study had not been so convincing, plenty of people would not “know” the truth about race and death sentencing—not all along, not now, not ever. Undeniable facts on the ground alone would not have produced a clean sweep of Justices of the Supreme Court. The Court is perfectly capable of looking facts in the eye and denying their existence.99 In this case, only a minority of the Justices directly acknowledged the truth of the Baldus study, but none of them has ever tried to deny it. It would have been a losing battle.

No single accomplishment, however impressive, does justice to a person as remarkable as David Baldus. What Baldus achieved in McCleskey, however, is worth dwelling on, and not only because of its historic importance. David Baldus forced reluctant judges to face up to facts they would have preferred to ignore. He was simply too hard-working and open and thorough to be doubted. This came through in the record. It was reinforced by the evaluations of other renowned scholars. And it was solidified over time by the impressive body of work he continued to produce until his untimely death.100

99. United States v. Armstrong, 517 U.S. 456 (1996), for example, addressed a related issue. Armstrong presented evidence that all federal crack defendants in the federal district in which he was prosecuted were black, but that many—perhaps most—state crack defendants were white. The evidence was sketchy, but that was to be expected: he was merely asking for discovery, to get information from the government to attempt to prove a claim of racial discrimination in federal charging for trafficking in crack cocaine. The standard the defendant must meet to obtain discovery is supposed to be low; he only has to produce “some evidence” that similarly situated people of a different race are not being prosecuted. The district court found that he had met that standard, as he plainly had, and the circuit court affirmed. But Chief Justice Rehnquist, writing for six members of the Court, concluded that somehow this showing was not even “some evidence” of selective prosecution. According to the Supreme Court, the disturbing facts Armstrong had assembled were worth nothing. 100. From the McCleskey decision until his death in June 2011, Baldus published many articles and reports on the administration of the death penalty, and conducted at least half a dozen major studies on the topic, some of which were as important and innovative as his better-known studies in Georgia. E.g., DAVID C. BALDUS, DEATH PENALTY PROPORTIONALITY REVIEW

PROJECT: FINAL REPORT TO THE NEW JERSEY SUPREME COURT (1991); David C. Baldus et al., Arbitrariness and Discrimination in the Administration of the Death Penalty: A Legal and Empirical Analysis of the Nebraska Experience (1973–1999), 81 NEB. L. REV. 486 (2002); David C. Baldus et al., Race Discrimination and the Death Penalty in the Post-Furman Era, 83 CORNELL L. REV. 1638 (1998); David C. Baldus et al., Racial Discrimination in the Administration of the Death Penalty: The



In a world in which “expert” is often synonymous with “partisan,” Baldus was known as a source of honest truth. Dave Baldus’s achievement in McCleskey is as much as anything a testament to his character—that of a tireless, selfless, passionate, inquisitive scientist.

Experience of the United States Armed Forces (1984–2005), 101 J. CRIM. L. & CRIMINOLOGY 1227 (2011).

david baldus and the legacy of mccleskey v. kemp

Documents